Growth factors have a central role in the progression of primary melanoma cells of the biologically early radial growth phase (RGP) to the tumorigenic vertical growth phase (VGP) with competence for metastasis. Factors produced by VGP primary and metastatic melanoma cells stimulate not only self in an autocrine circuit, but also fibroblasts and endothelial cells in the tumor stroma. There is a strong positive feedback from activated fibroblasts for tumor survival and growth and the first aim investigates the mechanisms of how fibroblasts shape the microenvironment of the tumor. It will be tested (Aim 1) whether IGF-1, derived from fibroblasts, acts as an essential growth and survival factor for biologically early melanoma cells, followed by the evaluation of a possible rote of fibroblasts in the differentiation of endothelial cells. Only fibroblasts and not melanoma cells can trigger capillary network formation of endothelial cells despite secretion of angiogenic growth factors by both normal and malignant cells. These studies will define the significance of stromal fibroblasts for melanoma progression and should determine whether this cell type can be targeted for tumor therapy in future studies. Overexpression of growth factors in dermal fibroblasts of human skin in combination with UV irradiation leads to melanocyte transformation and the development of lesions that resemble invasive melanoma in patients. Using first human embryonic stem cells and then stem cells from bone marrow and skin, the hypothesis is tested that fibroblasts control melanocyte stem cell differentiation and transformation (Aim 2). Dermal fibroblasts play an essential role in maintaining a homeostatic balance of differentiation and growth in the epidermis, and the mechanisms of homeostatic control ofmelanocytes and their dysregulation by the dermal microenvironment will be investigated. Using an in vitro in vivo transformation model of artificial human skin grafted to immunodeficient mice, the hypothesis that melanocyte precursor cells are more prone to transformation than mature melanocytes will be tested. These studies will determine whether melanocyte precursor cells can give rise to melanomas and how dermal fibroblasts contribute to their transformation. From metastatic melanomas, subpopulations are isolated and characterized to test the hypothesis that melanoma stem cells show similarities with melanocyte stem cells and require specific targeting for therapy.